Chapter 2. Electricity Sector Reform: Lessons and Policy Options13
- Trevor Alleyne, and Mumtaz Hussain
- Published Date:
- August 2013
SSA faces chronic power problems, including insufficient generation capacity, low access, poor reliability, and high costs and tariffs. The combined power generation capacity of the 48 SSA countries is about 80 gigawatts, less than that of Spain. Less than 3 in every 10 Africans have access to electricity. Per capita consumption of electricity is extremely low, averaging 40 kWh a month and only 10 kWh if South Africa is excluded. Power is unreliable: 15 percent of installed capacity is not operational because of lack of maintenance of aging equipment, and power outages are frequent. As a result, expensive own generation constitutes a significant portion of total installed capacity. In the Democratic Republic of the Congo and Equatorial Guinea, backup generators account for half of installed capacity. For West Africa as a whole, back-up generators account for 17 percent of installed capacity. Notwithstanding limited and unreliable supply, power is expensive: the average tariff in SSA is $0.17 per kWh, about twice that in other developing countries (Eberhard and Shkaratan, 2012).
The immediate reason for such shortfalls is underinvestment in the power sector. In 1974–2008, per capita production of electricity in SSA increased only marginally, lagging developments in other developing countries (Figure 18). As a result, although in the mid-1970s power supply in SSA compared favorably with most developing countries, by 2009 it ranked below all other regional groupings.
Figure 18.Sub-Saharan Africa: Electricity Production Compared to Other Regions, 1975–2009
Source: World Development Indicators; and IMF staff estimates.
Power supply in SSA is not only more limited, but tariffs are also higher than in other developing countries (Figure 19). This is mainly due to high costs of producing energy in SSA (Figure 20). Most countries rely on small generation units (that do not benefit from economies of scale) and operate expensive thermal plants, often using heavy fuel oils or diesel rather than cheaper natural gas.
Figure 19.Sub-Saharan Africa: Residential Tariffs Compared to Other Regions1 Figure 20.Sub-Sarahan Africa: Cost Factor for Residential Tariffs, Compared to Other Regions1
Underinvestment in power is, in part, related to subsidies for electricity consumption that have been largely borne by power utilities and prevented cost recovery. As described in Chapter 1, this has been a disincentive for new private sector investment. Insufficient cost recovery has also reduced the capacity of state-owned utilities to properly maintain plants and equipment and has left virtually no resources to expand operations and adequately address growing demand.
The objective of this chapter is to draw policy lessons from power sector reforms in Africa. These typically aim at multiple objectives: reducing budgetary costs of energy subsidies; increasing energy supply to meet excess demand; expanding access to foster inclusive economic growth; and making energy more affordable.
Although tariff changes grab headlines, addressing the problem of underinvestment is also heavily dependent on other reforms in the sector. Indeed, given that tariffs are already high in SSA, improving cost recovery for power providers will need to lean more heavily on reducing costs and improving efficiency. The extent of SSA’s power crisis demands tackling several policy and institutional challenges to improve the sector’s performance and financing. Such challenges include (1) strengthening sector planning; (2) reenergizing reform of public utilities to enhance their technical and operational efficiency; (3) improving access; and (4) expanding regional trade in power. A holistic approach to electricity sector reform, as applied in countries like Kenya and Uganda, can result in important payoffs in terms of increased power supply, expanded access, and enhanced financial sustainability of electricity enterprises (Box 2).
These challenges are all interrelated and must be dealt with simultaneously. Instead, some countries have had piecemeal approaches or fallen into a trap of false trade-offs, for example, between cost recovery (i.e., the need to increase tariffs) and affordability (i.e., the need to expand access to all). The utilities and the authorities can take several actions to lower subsidy costs and enhance cost recovery while promoting access and enhancing sector efficiency. We explore these actions in the following sections.
Box 2:Energy Reforms Pay Off in Kenya and Uganda
In early 2000s, both Kenya and Uganda implemented a multitude of reforms aimed at improving performance of the power sector.
- In Kenya, reform efforts culminated in a new energy policy in 2004, substantial increase in power tariffs in 2005 to reflect long-run marginal costs, introduction of an automatic pass-through mechanism to adjust tariffs for changes in fuel costs, and reconstitution of the Electricity Regulatory Commission.
- In Uganda, electricity sector reform included the passage of a new Electricity Act (1999); the establishment of a regulatory agency (2000); and the unbundling of the power utility (2001) and concessioning of its parts (2003–05). In 2006, power tariffs were almost doubled, raising the average effective tariff to US$.018 per kWh to reflect long-run marginal costs of power.
In both countries, the reforms led to improvements in the electricity sector. Since mid-2000s, power generation increased steadily, distribution losses declined, and the number of customers served by grid-supplied power increased substantially.
- Power supply increased. The private sector’s involvement in power generation combined with increased tariffs led to a substantial boost in power supply (see table). In the posttariff increase period, average annual increase in power supply in Kenya was over 5 percent and in Uganda over 9 percent.
- Distributional efficiency improved. Distribution losses of power have steadily fallen and bill collection rates improved. In Kenya, line losses declined from 18 percent in 2005 to 16 percent in 2011, and the collection rates increased from 85 percent of total power bills in 2005 to 99 percent in 2011. Efficiency gains were even stronger in Uganda: distribution losses declined from 38 percent in 2005 to about 27 percent in 2011, and collection rates increased from 80 percent of total power bills in 2005 to 95 percent in 2011.
- Access to grid-supplied power expanded. After limited progress early on, the number of customers with access to grid-supplied power in Uganda increased by 41 percent between 2006 and 2011. In Kenya, access increased by nearly 140 percent between 2005 and 2011.
- Progress on reducing quasi-fiscal costs was mixed. In Kenya, tariff increases in 2005 combined with the automatic price adjustment mechanism, and improved efficiency helped eliminate quasi-fiscal costs by 2009. In Uganda, notwithstanding efficiency gains, the quasi-fiscal deficit of power sector increased steady until 2011 because of higher fuel costs and lack of adjustments in power tariffs. In January 2012, however, tariffs were raised to the cost recovery levels and a pass-through mechanism to adjust tariffs in response to variation in generation costs is being developed.
|Electricity supplied (GWh)||1,741||1,503||2,387||2,477||59||9.3|
|Electricity billed (GWh)||1,075||990||1,732||1,886||75||11.2|
|Distribution losses (percent of total power)||38.3||34.1||27.5||n.a.||−20||−4.4|
|Collection rate (percent of total bills)||80.0||84.0||95.3||n.a.||13||2.5|
|Number of customers (in thousands)||292||298||420||459||41||6.8|
|Of which: industrial customers (in thousand)||n.a.||1.0||1.8||2.1||82||12.0|
|Quasi-fiscal costs (percent of GDP)||1.6||1.9||2.6||0.7||…||…|
|Electricity supplied (GWh)||5347||5697||7303||n.a.||37||5.2|
|Electricity billed (GWh)||4379||4580||6123||n.a.||40||5.6|
|Distribution losses (percent of total power)||18.1||19.6||16.2||n.a.||−11||−3.9|
|Number of customers (in thousands)||735||802||1753||n.a.||139||14.5|
|Quasi-fiscal costs (percent of GDP)||0.6||0.6||02||n.a.||…||…|
For Kenya, the changes are calculated between 2005 (major reform year) and 2011; for Uganda, the changes are calculated between 2006 (major reform year) and 2011.
Data is for 2009.
For Kenya, the changes are calculated between 2005 (major reform year) and 2011; for Uganda, the changes are calculated between 2006 (major reform year) and 2011.
Data is for 2009.
It is important to note that each country is different, and the appropriate reforms will be a function of the country’s characteristics. For countries with much higher costs owing, for example, to geography, small thermal systems, or small population size, reforms may need to focus on regional solutions. For example, in Burkina Faso, the authorities plan to focus their investment on transmission to increase network capacity through imports. If the main issue is the extent of technical or commercial losses (e.g., Ghana or Sierra Leone), actions should target the engineering and commercial aspects of service delivery. If the main issue is exposure to weather shocks (if the country relies heavily on hydropower, e.g., Uganda), then “reenergizing” power pools would certainly figure prominently in any strategy to bring production costs down—or reduce outages. Also, any country relying on hydropower will have to invest in transmission to allow heavy seasonal power loads to be efficiently routed through the system.
Power Planning and Institutional Structure
Power planning is essential for successful reforms in the electricity sector, notably, lower costs and enhanced access and service quality. Much reform in SSA’s power sector has been piecemeal, lacking a comprehensive plan. Power planning is the process of projecting the yearly energy balance of a country in the medium to long term to optimize the development of the sector. In a nutshell, power planning explores least-cost options to meet projected demand. The planning process enables policymakers to focus on bringing down the long-run marginal cost via the cheapest option, be it domestic or external (e.g., participation in power pools). For most SSA countries, a critical issue is to identify economies of scale.
Effective planning involves strategic decisions in a number of areas. Important decisions cover (1) the domestic production mix (fuel oil, coal, hydro, etc.); (2) private sector participation in generation, which depends on pricing policy and regulatory capacity; (3) load planning: harnessing supply during the rainy season and providing alternative supply during the dry season; (4) financial planning: quantifying financing gaps and identifying the mix of financing, including donors; (5) off-grid options for expanding access in areas where expansion of the grid is prohibitively costly; and (6) regional solutions, including participation in power pools or enhancing integration in regional markets.
The choice of institutional organization should be one that creates incentives to enhance efficiency, given the country’s specific conditions.
The focus of early electricity reforms in SSA was on unbundling and privatization, but this has not been a panacea.14 Unbundling has the advantages of separating the natural monopolies in transmission and distribution from the naturally competitive stage of energy generation. But unbundling vertically integrated energy utilities only makes economic sense for countries large enough to support multiple generators operating at an efficient scale, which excludes most countries in SSA (Besant-Jones, 2006).15
Thus, a strong case can be made for a hybrid model. Countries with excess energy demand—a standard condition in SSA—should consider increasing installed capacity by removing the monopoly power from generation and creating incentives for private participation. In fact, hybrid power markets—with the incumbent state-owned utility acting as the single buyer of electricity from Independent Power Producers (IPPs)—have become the most common industry structure in SSA.16 This approach is appealing particularly for LICs and fragile states, where the institutional and organizational changes required by unbundling would stretch thin local capacities. A key issue in hybrid markets is how to foster competition and manage system growth. The prevailing single-purchaser model could be revised to allow more flexibility for IPPs to sell directly to large customers as well as to the national utility.
Enhancing Efficiency, Reducing Costs, and Raising Revenue
Enhancing the efficiency of utilities should be a priority, preferably to be contemplated before tariffs are increased. As described in Chapter 1, electricity companies in SSA typically present huge inefficiencies, both in terms of technical and commercial losses (see below). In the median utility, payment is received for only half of all electricity generated (Eberhard and others, 2008). The main message is that utilities should define action plans focused on achieving sustainable quality in electricity supply, reducing losses, and increasing collection rates.
Improving Operational Efficiency
A priority action for electricity utilities is to put in place information management systems (IMS) to facilitate management of business operations and monitoring the performance of electricity companies. IMS are used by several state-owned electricity companies in SSA, with Kenya Power and Lighting Company (KPLC) taking the lead.17 There are no significant technical or commercial barriers to the introduction of IMS. The main barrier arises from the strong reluctance of managers and staff of monopolistic utilities. A lack of transparency in management and operations is the main factor allowing corruption and operational inefficiencies (e.g., overstaffing).
Tariff regulation can be designed to establish incentives to improve efficiency of public and private electricity distribution companies. Given the already high tariffs in SSA, it is important to ensure that tariff increases are not used to cover up and perpetuate inefficient business practices. Multiyear tariffs can be based on revenue inclusive of an allowance for losses. If the company manages to operate with lower losses than allowed in the formula, it keeps the difference as an additional profit until the next tariff review.18 If the company fails to achieve the loss reduction, the gap between the allowed and the actual losses has to be covered by the company. This system has been used extensively by reforming countries in Latin America and the Caribbean and was extremely effective in Argentina, Chile, and El Salvador (Antmann, 2009). Several countries in sub-Saharan Africa, such as Cameroon, Uganda, and Kenya, have also adopted such a regulatory regime.
Reducing Costs and Demand Management
Technical and commercial losses, which are important components of costs, are usually under the control of the management of utilities. Technical losses are an engineering issue and consist mainly of power dissipating through transmission and distribution lines, transformers, and measurement systems. Commercial losses are caused by theft, nonpayment, and errors in accounting and record keeping. Thus, they mainly result in loss of revenue (see below).
Technical losses are difficult to address without adequate and centrally monitorable energy mapping—i.e., the measurement of the energy flows in the system. Although IMS are an integral part of mapping energy use, in undermaintained and underinvested systems or when new investments come on board, mapping might require new investments. These investments tend to enjoy high internal rates of return and enhance service quality, including by reducing loan shedding. The improvement in service quality saves money to businesses and households and establishes a customer base with a stake in the improved efficiency of the utility.
An integral part of any strategy to reduce operation costs is demand management to maximize efficiency in electricity supply. Promotion of energy-saving solutions for consumers can have a significant positive effect. For example, utilities in SSA are providing free compact fluorescent bulbs (CFLs), which have helped reduce demand and costs in Cape Verde, Ethiopia, Malawi, Uganda, and Rwanda.19 Also, nongrid options often can provide cheaper and faster alternative solutions to expanding access through grid expansions (see below). The human geography of SSA, with large percentages of the population living in rural areas—often in small settlements—makes universal access through grid expansion prohibitively expensive. However, for nongrid options to succeed, they need to be integrated in power planning, and a supportive regulatory environment for small-scale off-grid operators must be established.
A revenue recovery and protection plan (RRP) should be a primary tool to improve the utility’s financials. The plan should first target “high-value” large customers, ensuring that all the energy consumed by this segment is metered, billed, and collected in a sustainable manner.20 The RRP requires investment in metering and communication devices and in software to process and analyze the data and, importantly, the creation of a specialized unit within the utility to manage the project, staffed with skilled personnel with high integrity.21 The systems and procedures developed for “high-value” customers should then be adapted and gradually extended to medium- and low-value customers (Antmann, 2009). In this context, prepaid electricity has proven a useful tool to promote payment discipline and improve collection rates in low-income customer segments (e.g., Nigeria, Kenya, Uganda, Senegal).
At low capacity, particularly in fragile states, bill collection could be outsourced as a concession.22 Collection is a distinct function from metering and billing, and utility companies may not be particularly equipped for it. For example, the National Power Authority of Sierra Leone issued a concession to a local commercial bank, resulting in higher collection rates in 2011 and avoiding the hiring of collection staff.
Tariff Design, Changes, and Targeting
Tariff design often has to meet competing objectives. Sound pricing of power is critical to meeting the huge investment needs of the sector. At the same time, policymakers need to ensure they meet equity and affordability objectives. Marrying all these goals is challenging, and few countries have been able to achieve all these objectives simultaneously. Chad, Mozambique, Rwanda, and Uganda have done well on cost recovery but poorly on affordability and equity, while South Africa, the Democratic Republic of Congo, Tanzania, and Zambia have fared well on the social objectives but have not been able to achieve cost recovery (Briceño-Garmendia and Shkaratan, 2011). Nevertheless, other country experiences, such as the progress in Kenya, indicate that it is possible to make substantial progress in both cost recovery and affordability.
There is considerable scope to achieve subsidy savings without compromising targeting. The most common electricity tariff regime is based on consumption (increasing block tariffs or IBTs), where consumers face higher unit prices on higher blocks of consumption.23 However, IBT regimes have been implemented with a variety of flaws. In many cases, they tend to be highly regressive because consumption in the first block is subsidized even for those with higher total consumption and income. In addition, although in some countries the size of the first—or lifeline—block seems reasonable given subsistence consumption (e.g., 15 kWh in Uganda), in others it would appear very large (e.g., 300 kWh in Ghana and Zambia) (Briceño-Garmendia and Shkaratan, 2011). In some countries, block prices increase too slowly with higher volumes. Thus, cost recovery is compromised even for higher blocks, and better-off households benefit from the subsidies. Some countries also duplicate the lifeline block with social tariffs. A more efficient and progressive design is the volume-differentiated tariff (VDT) where consumption above a threshold leads to a higher price on all consumption. The VDT is an effective method to efficiently target lifeline blocks, thus reducing costs associated with subsidy schemes for the poorest (e.g., Cape Verde). This requires progress in metering, which remains an important challenge in most SSA countries. Regardless of pricing mechanism, correct calibration of block sizes and associated price levels requires a good knowledge of consumption patterns derived from Household Expenditure Surveys. More broadly, there are alternatives to consumption-based targeting that perform considerably better, such as geographic targeting (e.g., Liberia) or means testing.
More recently, emphasis has moved from subsidizing tariffs to subsidizing connections. Surveys show that prohibitive connection costs (e.g., around $1,000 in Liberia) are the main factor preventing people from accessing grids. Meanwhile, “willingness to pay” analysis confirms that households are often willing to pay for electricity, because they would be saving the expense of alternative and less convenient energy sources (e.g., Ethiopia). Subsidization can take different forms, including interest-free loans (e.g., Kenya) or deferred payments by installments (e.g., Liberia), broadly matching the savings from switching from expensive energy sources to the grid.24 In fact, expanding the customer base in areas close to the main lines can make financial sense for utilities, because it may improve the ratio of paying to nonpaying customers with limited infrastructure investment (e.g., Liberia, Kenya).
Other issues in tariff design merit attention. Special pricing of power to commercial and industrial consumers is important given they often account for one half or more of revenue. Special tariffs are sometimes provided to large electricity users, such as large industrial and mining customers, and are not reflected in the general tariff structure, or even in estimates of the subsidy (e.g., Copperbelt Energy Corporation in Zambia). Originally intended to guarantee minimum demand to support the development of large power projects, they now impose a significant fiscal cost given competing demands, some from more competitive enterprises.
Only 30 percent of the population in SSA is connected to the grid (International Finance Corporation, 2012). Africa is thus home to the world’s largest off-grid population: approximately 590 million people and more than 10 million microenterprises have no connection to their national electric grid (International Energy Agency, 2011). Power plans must thus deal with the unavoidable: universal access to the grid is decades away for most countries in SSA. Large and dispersed rural populations make grid expansion costs prohibitive. But coverage in urban areas is also limited, particularly in poor neighborhoods, often because of affordability issues.
Over the last decade, several countries have established special-purpose agencies and funds for rural electrification. On average, greater progress in access in rural areas has been made in countries with rural electrification agencies, especially if supported by dedicated funds.25 Countries with higher urban populations also tend to have higher levels of rural electrification, because urban populations tend to subsidize rural electrification.
Several SSA countries have implemented rural electrification programs to enhance access outside the main cities (e.g., Mali, Box 3; Kenya, Ethiopia, Liberia, Rwanda, Uganda). Given SSA’s demographics, expanding the grid to rural areas would be prohibitive. However, greater access to power is still a key to human and economic development in rural areas. Thus, several countries have set up program and institutions providing alternative solutions to power supply.
Box 3.Mali: Rural Electrification Program Succeeds in Expanding Access
Mali succeeded in increasing rural electrification through off-grid solutions and private sector participation. In rural areas of Mali, only around 13 percent of the population could access electricity in 2009. Most rural households thus meet their lighting and small power needs with kerosene, dry cells, and car batteries. More than 80 percent of Malians use wood or charcoal for cooking and heating. These sources of energy cost about $1.5/kWh, more than 10 times the price from the grid. To address these problems, the government established a rural electrification agency and a Rural Electrification Fund aimed at providing partial start-up capital for private operators of mini-grids. The project fostered local private sector participation. As of May 15, 2010, 43,311 off-grid connections for households and public lighting provided electricity to about 650,000 people. In addition, about 803 public institutions, including 172 schools and 139 health centers, received off-grid access. With the installation of multifunctional platforms by local operators in 64 communities, resulting in 7,200 connections as of mid-2011, numerous business opportunities were created. The platforms are diesel motors that combine electricity production with other services such as milling, husking, pumping water, charging batteries, running lights, and powering tools.26 The electrification program also fostered the use of renewable energy: in six years, more than 7,926 households and 500 institutions were connected to solar systems (Eberhard and others, 2011; World Bank, various years).
Regional Electricity Trade
Cross-border trade in power has the potential to considerably reduce the cost of energy supply (Figure 21). Depending on the country and its neighbors, the cost of kWh could be reduced from US$0.01 to US$0.07 by importing power at prices below the domestic cost of production. However, the gains from trade could be much larger, because exporting countries could exploit economies of scale and importing countries could abandon expensive small-scale options.
Figure 21.Sub-Saharan Africa: Potential Savings from Cross-Border Power Trade
Source: Foster and Briceño-Garmendia (2010).
The potential for trade is large, because resources for energy generation are unevenly distributed. Oil and gas reserves are in the Gulf of Guinea and Sudan, Ethiopia, Chad, Mozambique, Namibia, and Tanzania; hydropower mostly in the Democratic Republic of the Congo and Ethiopia; coal deposits in South Africa; geothermal energy in Kenya, Ethiopia, and Djibouti; and wind power potential in Southern Africa.
However, the 43 countries participating in the four African power pools have generated less trade than initially hoped.27 Although a limiting factor has been that few African countries have excess supply to trade, power pools could be more successful with increased investments in grid interconnections, a legal framework for cross-border electricity exchange, and mechanisms for dispute resolution.